SOC 4930 & SOC 5050 - Week 02, Lecture 02b

Lecture slides for Week 02, Lecture 02b of the Saint Louis University Course Quantitative Analysis: Applied Inferential Statistics. These slides cover the basics of the R package ggplot2.

Transcript

1. THE GRAMMAR OF 
 GRAPHICS QUANTITATIVE ANALYSIS CHRISTOPHER PRENER, PHD FALL, 2017 WEEK 02 LECTURE 02B
2. AGENDA QUANTITATIVE ANALYSIS / WEEK 02 / LECTURE 02B 1. Statistical Transformations 2. Aesthetic Adjustments 3. Coordinate Systems 4. The Grammar of Graphics
3. STATISTICAL TRANSFORMATIONS 1
4. 1. STATISTICAL TRANSFORMATIONS BASIC BAR PLOT
5. UNDER THE HOOD stat_count() geom_bar()
6. ggplot2::ggplot(data = dataFrame) + geom_bar(mapping = aes(aesthetic)) Example - the mpg data from ggplot2: ggplot(data = mpg) + geom_bar(mapping = aes(class)) 1. STATISTICAL TRANSFORMATIONS BAR PLOTS
7. 1. STATISTICAL TRANSFORMATIONS BAR PLOTS ggplot2::ggplot(data = dataFrame) + geom_bar(mapping = aes(aesthetic), stat = “identity”) Example - the mpg data from ggplot2: ggplot(data = mpg) + geom_bar(mapping = aes(class), stat = “identity”) The stat “identity” is implied by default when you use the bar geom
8. 1. STATISTICAL TRANSFORMATIONS CONTINUOUS VARIABLES stat_bin()
9. 1. STATISTICAL TRANSFORMATIONS LINE PLOTS stat_bin()
10. 1. STATISTICAL TRANSFORMATIONS AREA PLOTS ggplot2::ggplot(data = dataFrame) + geom_area(mapping = aes(aesthetic), stat = “bin”) Example - the mpg data from ggplot2: ggplot(data = mpg) + geom_area(mapping = aes(hwy), stat = “bin”) This geom is for use with one continuous variable, stat must be included
11. 1. STATISTICAL TRANSFORMATIONS AREA PLOTS
12. 1. STATISTICAL TRANSFORMATIONS BOX PLOTS
13. 1. STATISTICAL TRANSFORMATIONS SUMMARIZING VALUES Example - the mpg data from ggplot2: ggplot(data = mpg) + stat_summary( mapping = aes(x = class, y = hwy), fun.ymin = min, fun.ymax = max, fun.y = median )
14. 1. STATISTICAL TRANSFORMATIONS SUMMARIZING VALUES
15. AESTHETIC ADJUSTMENTS 2
16. ggplot2::ggplot(data = dataFrame) + geom_point(mapping = aes(x = var1, y = var2), color = “color”) Example - the mpg data from ggplot2: ggplot(data = mpg) + geom_point(mapping = aes(x = displ, y = hwy), color = “blue”) You cannot use the fill option instead of color with this geom 2. AESTHETIC ADJUSTMENTS ADDING COLOR ARBITRARILY
17. 2. AESTHETIC ADJUSTMENTS ADDING COLOR ARBITRARILY
18. ggplot2::ggplot(data = dataFrame) + geom_point(mapping = aes(x = var1, y = var2), color = “color”) Example - the mpg data from ggplot2: ggplot(data = mpg) + geom_point(mapping = aes(x = displ, y = hwy), color = “blue”) You cannot use the fill option instead of color with this geom For now, avoid setting colors arbitrarily and let ggplot2 do the work instead! 2. AESTHETIC ADJUSTMENTS ADDING COLOR ARBITRARILY
19. ggplot2::ggplot(data = dataFrame) + geom_point(mapping = aes(x = var1, y = var2, color = “color”)) Example - the mpg data from ggplot2: ggplot(data = mpg) + geom_point(mapping = aes(x = displ, y = hwy, color = “class”)) You cannot use the fill option instead of color with this geom 2. AESTHETIC ADJUSTMENTS ADDING COLOR TO THE AESTHETIC
20. 2. AESTHETIC ADJUSTMENTS ADDING COLOR TO THE AESTHETIC
21. ggplot2::ggplot(data = dataFrame) + geom_point(mapping = aes(x = var1, y = var2, color = var3)) Example - the mpg data from ggplot2: ggplot(data = mpg) + geom_point(mapping = aes(x = displ, y = hwy, color = class)) You cannot use the fill option instead of color here 2. AESTHETIC ADJUSTMENTS ADDING COLOR BASED ON OTHER VALUES
22. 2. AESTHETIC ADJUSTMENTS ADDING COLOR BASED ON OTHER VALUES
23. ggplot2::ggplot(data = dataFrame) + geom_bar(mapping = aes(x = var, fill = var)) Example - the mpg data from ggplot2: ggplot(data = mpg) + geom_bar(mapping = aes(x = class, fill = class)) This can also be done with the color option, but fill is preferable 2. AESTHETIC ADJUSTMENTS ADDING COLOR BASED ON OTHER VALUES
24. 2. AESTHETIC ADJUSTMENTS ADDING COLOR BASED ON OTHER VALUES
25. ggplot2::ggplot(data = dataFrame) + geom_bar(mapping = aes(x = var, fill = var)) Example - the mpg data from ggplot2: ggplot(data = mpg) + geom_bar(mapping = aes(x = class, fill = trans)) 2. AESTHETIC ADJUSTMENTS ADDING COLOR BASED ON OTHER VALUES
26. 2. AESTHETIC ADJUSTMENTS ADDING COLOR BASED ON OTHER VALUES
27. ggplot2::ggplot(data = dataFrame) + geom_bar(mapping = aes(x = var, fill = var), position = “dodge”) Example - the mpg data from ggplot2: ggplot(data = mpg) + geom_bar(mapping = aes(x = class, fill = trans), position = “dodge”) 2. AESTHETIC ADJUSTMENTS POSITION ADJUSTMENTS - DODGE
28. 2. AESTHETIC ADJUSTMENTS POSITION ADJUSTMENTS - DODGE
29. 2. AESTHETIC ADJUSTMENTS “MISSING” POINTS
30. ggplot2::ggplot(data = dataFrame) + geom_point(mapping = aes(x = var1, y = var2), position = “jitter”) Example - the mpg data from ggplot2: ggplot(data = mpg) + geom_point(mapping = aes(x = displ, y = hwy), position = “jitter”) 2. AESTHETIC ADJUSTMENTS POSITION ADJUSTMENTS - JITTER
31. 2. AESTHETIC ADJUSTMENTS POSITION ADJUSTMENTS - JITTER
32. COORDINATE SYSTEMS 3
33. 3. COORDINATE SYSTEMS ▸ French/Dutch philosopher and mathematician ▸ Published the idea of data arrayed along an axes in 1647 ▸ Has been transformed into what we now understand to be Cartesian coordinate systems ▸ A number of other thinkers wrote about similar ideas independently in the same period RENÉ DESCARTES
34. 3. COORDINATE SYSTEMS CARTESIAN COORDINATE SYSTEMS
35. 3. COORDINATE SYSTEMS CARTESIAN COORDINATE SYSTEMS
36. 3. COORDINATE SYSTEMS FLIPPING COORDINATES ggplot2::ggplot(data = dataFrame) + geom_bar(mapping = aes(x = var, fill = var) + coord_flip() Example - the mpg data from ggplot2: ggplot(data = mpg) + geom_bar(mapping = aes(x = class, fill = trans), position = “dodge”) + coord_flip()
37. 3. COORDINATE SYSTEMS FLIPPING COORDINATES
38. THE GRAMMAR OF GRAPHICS 4
39. 4. THE GRAMMAR OF GRAPHICS A BASIC TEMPLATE ggplot2::ggplot(data = dataFrame) + geom(mapping = aes(aesthetics), stat = statistics, position = position, ) + coordinateFunction + facetFunction